Search found 409 matches

by v-zero
Thu Apr 25, 2019 9:59 pm
Forum: APBRmetrics
Topic: Reconstructing Box Plus/Minus
Replies: 105
Views: 48196

Re: Reconstructing Box Plus/Minus

That makes a lot of sense, Colts18. I would like to somehow do things that way, but it certainly adds a lot of computational complexity. For many leagues, that data will not be readily available. I also don't think it can actually work if the metric is nonlinear. For a linear metric, it would be a ...
by v-zero
Mon Jul 03, 2017 7:16 pm
Forum: APBRmetrics
Topic: Way-too-early 2017-18 Forecasts
Replies: 39
Views: 31840

Re: Way-too-early 2017-18 Forecasts

Based on what has gone on so far, OKC looks to me like a 54 win team, Houston a 60 win team, Denver a 50 win team and Minnesota a 50 win team.
by v-zero
Tue Mar 07, 2017 7:12 pm
Forum: APBRmetrics
Topic: Is Westbrook Exposing BPM ?
Replies: 39
Views: 24578

Re: Is Westbrook Exposing BPM ?

As for improvements, I'd love to hear ideas. Perhaps it should be a sqrt(USG x AST) instead of its current form? I have begun work on a new version of BPM to address a few issues I've seen and to take advantage of better long-term RAPM. In my experience the only way to achieve a stable player model...
by v-zero
Thu Feb 16, 2017 7:44 pm
Forum: APBRmetrics
Topic: p-hacking
Replies: 14
Views: 10411

Re: p-hacking

This is why when considering a lot of variables and possible interaction variables then application of AIC or cross-validation is far more valuable for discerning actual value.
by v-zero
Fri Apr 15, 2016 2:08 pm
Forum: APBRmetrics
Topic: Running Regressions On Pace
Replies: 9
Views: 8316

Re: Running Regressions On Pace

From a sort of theoretical perspective the best way to average the pace of two teams is to invert them, then average them, then invert that average. This is because the inverse of pace is time, and each team plays the same number of possessions (more or less) so what you really want to know is how l...
by v-zero
Thu Jan 07, 2016 9:45 am
Forum: APBRmetrics
Topic: Cumulative lineup height RAPM
Replies: 13
Views: 21187

Re: Cumulative lineup height RAPM

This done using the geometric mean of heights rather than the cumulative heights would be really interesting, not that it isn't interesting as is.
by v-zero
Mon Mar 23, 2015 2:30 pm
Forum: APBRmetrics
Topic: Iverson's career stats
Replies: 100
Views: 73750

Re: Iverson's career stats

Players like guys who gave everything they had, who were competitors above all else, and who fought adversity. That's why players love Iverson so much.
by v-zero
Wed Feb 11, 2015 7:27 pm
Forum: APBRmetrics
Topic: Proof of Diminishing returns on Rebounds, USG-EFF, and more
Replies: 28
Views: 21833

Re: Proof of Diminishing returns on Rebounds, USG-EFF, and m

Was anything done to account for selection bias? If not then that probably explains the odd age values.
by v-zero
Sat Jan 24, 2015 1:11 pm
Forum: APBRmetrics
Topic: My side-project: recappd.com
Replies: 17
Views: 14473

Re: My side-project: recappd.com

Really cool, and similar to something I have planned (but not identical) for a while now. I'll definitely be using this when need a lightning-quick recap. :D
by v-zero
Wed Dec 31, 2014 12:38 pm
Forum: APBRmetrics
Topic: Problems with Linearity Assumptions
Replies: 11
Views: 5956

Re: Problems with Linearity Assumptions

Methods such as the use of boosted decision stumps (with linear regressions at their terminal nodes if required) can avoid the linearity question for the most part.
by v-zero
Thu Dec 11, 2014 2:52 pm
Forum: APBRmetrics
Topic: Continuation of one metric prediction test discussion
Replies: 32
Views: 10073

Re: Continuation of one metric prediction test discussion

The use of ensemble models has long been a feature of the machine learning community. I suggest anybody wanting to learn about the power of 'blends' do lots of reading in that area, and forget about basketball specific research.
by v-zero
Wed Dec 10, 2014 2:22 pm
Forum: APBRmetrics
Topic: Creating power/schedule adjusted rankings
Replies: 12
Views: 10504

Re: Creating power/schedule adjusted rankings

I keep seeing HCA mentioned in many of these, and yet can't find a single place mentioning how it is being calculated. Is HCA different than simply MOV while at home? or maybe like HCA = (AvgMOV-Home - AvgMOV-Away) / Games Played? Slight adjustment to what DSMok1 has said... If using a league avera...
by v-zero
Sun Dec 07, 2014 12:59 pm
Forum: APBRmetrics
Topic: 2014-15 MAX Ratings
Replies: 42
Views: 17636

Re: 2014-15 MAX Ratings

2. Finding out which seasons are "in-sample" and which seasons are "out-of-sample" for a metric is incredibly simple. Do I really need to explain it? :) So how would you do this? To you what does in-sample mean? I'm willing to bet it doesn't mean to you what it does to me. I know that BPM blend (wh...
by v-zero
Thu Dec 04, 2014 9:16 am
Forum: APBRmetrics
Topic: 2014-15 MAX Ratings
Replies: 42
Views: 17636

Re: 2014-15 MAX Ratings

It is pretty clear to me that the only way we will get a 'fair' test is if each model is fully explained and the means by which to reproduce it is given to Neil. Neil would then have to manually reproduce each model from data not provided by the contestants. Anybody unwilling to provide such informa...
by v-zero
Thu Nov 27, 2014 6:31 pm
Forum: APBRmetrics
Topic: LRAPM
Replies: 8
Views: 3870

Re: LRAPM

J.E. I actually have a question about the playing from ahead factor...do you think it is valid to include as part of the model when intending to represent player value? Surely it is an attribute of each player (the extent to which they cause their team to 'coast' by being good or play catch up by be...